7.1 Validity:
Validty is the degree to which a test measures what it is supposed to measure.
Meter is used for the purpose of length while scale is used for the purpose measuring of measuring weight. If meter is used for measuring length, it will be valid because it is supposed to measured length and if it is used for measuring weight, it will not be valid.
Suppose a test is developed to measure the scientific concepts for a particular class, it will be valid for this specific purpose. It will not be valid for any other purpose. If some one tries to assess English comprehension from this test, then the test will not be valid.Tests are designed for a variety of purposes. A test designed for measuring achievement in the subject of Biology will not be valid for measuring the personality. In the same way, a test designed for measuring achievement in the subject of Biology for class 5 will not be valid for measuring achievement in the subject of Biology for class 8.
Types of Validity:
1. Content Validity:
"Content validity is the degree to which a test measures an intended content area."It is concerned with the ability of a test to cover all contents. If a test is designed to measure the concepts of Biology and the items of the test deal only with five out of ten chapters then the test will show poor content validity.
A test with good content validity covers all the contents.
Content validity is of prime importance for achievement tests.It is measured by expert judgement. It can not be computed by any formula and cannot be expressed quantitatively. Usually experts are asked to assess the content validity of a test. However developing a table of specification for item development ensures the content validity. A table of specification is a device which results in a balanced coverage of contents.
2. Construct Validity:
A construct is a non-observable trait such as learning, anxiety, creativity, curiosity, scientific attitude, intelligence which explains behavior. Construct validity is the degree to which a test measures an intended hypothetical construct.A construct cannot be seen, only its effect can be observed e.g., intelligence cannot be seen but we was observe that some students learn faster than others. To explain this difference, a theory of intelligence developed. It was hypothesized that intelligence is related to learning. Tests were developed to measure intelligence. Students having high IQs tend to learn faster.Construct validity involves testing hypothesis deduced from a theory concerning the construct. For example, if a theory of intelligence hypothesized that students with high IQs learn faster then students achieving high scores on a test designed to measure intelligence if indeed learn faster, it will be evidence in support of construct validity of a test.
Setps:
1.Theorr: Student with high IQ learn faster.
2. Administer an intelligence Test.
3. Administer a test to assess the speed of learning.
4.Correlate the two sets of scores.
5. High correlation will be a proof in favour of construct validity of test.
3. Concurrent Validity:
Concurrent validity is the degree to which the scores on a test are related to the scores on another already established test administered at the same time.Often a test is developed that claims to do the same job as other tests e.g., a paper and pencil test that renc does the same job as performance test, will be preferred. In the the same way a shorter test doing the same job as cases longer test, will be preferred. In these cases, the encil lest and concurrent validity of the paper and pencil test and shorter test need to be established.
Setps:
1.Administer the new test.
2. Administer already developed valid test to the same group at the same time or shortly thereafter.nas!
3. Correlate the two set of scores. 4. High correlation coefficient will indicate good concurrent validity and vice versa.
4. Predictive Validity:
Predictive validity is the degree to which a test can predict how well an individual will do in future.A mathematics aptitude test that have high predictive validity will accurately predict which students will do well in mathematics and which will not. Predictive validity is of prime importance for tests which are used for the purpose of admission or selection of students. If the admission or selection test will have high predictive validity then the students will get high scores in future programmes.
Steps:
1. Administer the test (Maths aptitude test)
2.Wait until behavior to be predicted occurs (Wait Until students have completed course in Maths)
3.Obtain scores in final exams.
4. Correlate the two set of scores.
5. High correlation coefficient will indicate high predictive validity of the test administered.